Mário J. Silva

Danielle Caled, Paula Carvalho, Mário J. Silva

Most corpora approach misinformation as a binary problem, classifying texts as real or fake. However, they fail to consider the diversity of existing textual genres and types, which present different properties usually associated with credibility. To address this problem, we created MINT, a comprehensive corpus of news articles collected from mainstream and independent Portuguese media sources, over a full year period. MINT includes five categories of content: hard news, opinion articles, soft news, satirical news, and conspiracy theories. This paper presents a set of linguistic metrics for characterization of the articles in each category, based on the analysis of an annotation initiative performed by online readers. The results show that (i) conspiracy theories and opinion articles present similar levels of subjectivity, and make use of fallacious arguments; (ii) irony and sarcasm are not only prevalent in satirical news, but also in conspiracy and opinion news articles; and (iii) hard news differ from soft news by resorting to more sources of information, and presenting a higher degree of objectivity.

Silvio Amir, Glen Coppersmith, Paula Carvalho et al.

Mental illnesses adversely affect a significant proportion of the population worldwide. However, the methods traditionally used for estimating and characterizing the prevalence of mental health conditions are time-consuming and expensive. Consequently, best-available estimates concerning the prevalence of mental health conditions are often years out of date. Automated approaches to supplement these survey methods with broad, aggregated information derived from social media content provides a potential means for near real-time estimates at scale. These may, in turn, provide grist for supporting, evaluating and iteratively improving upon public health programs and interventions. We propose a novel model for automated mental health status quantification that incorporates user embeddings. This builds upon recent work exploring representation learning methods that induce embeddings by leveraging social media post histories. Such embeddings capture latent characteristics of individuals (e.g., political leanings) and encode a soft notion of homophily. In this paper, we investigate whether user embeddings learned from twitter post histories encode information that correlates with mental health statuses. To this end, we estimated user embeddings for a set of users known to be affected by depression and post-traumatic stress disorder (PTSD), and for a set of demographically matched `control' users. We then evaluated these embeddings with respect to: (i) their ability to capture homophilic relations with respect to mental health status; and (ii) the performance of downstream mental health prediction models based on these features. Our experimental results demonstrate that the user embeddings capture similarities between users with respect to mental conditions, and are predictive of mental health.

Silvio Amir, Rámon Astudillo, Wang Ling et al.

Recent approaches for sentiment lexicon induction have capitalized on pre-trained word embeddings that capture latent semantic properties. However, embeddings obtained by optimizing performance of a given task (e.g. predicting contextual words) are sub-optimal for other applications. In this paper, we address this problem by exploiting task-specific representations, induced via embedding sub-space projection. This allows us to expand lexicons describing multiple semantic properties. For each property, our model jointly learns suitable representations and the concomitant predictor. Experiments conducted over multiple subjective lexicons, show that our model outperforms previous work and other baselines; even in low training data regimes. Furthermore, lexicon-based sentiment classifiers built on top of our lexicons outperform similar resources and yield performances comparable to those of supervised models.